Western Australia - WA Health _____ Title Injury in Western Australia: The Health System Costs of...
44
Injury in Western Australia The Health System Costs of Falls in Older Adults in Western Australia
Transcript of Western Australia - WA Health _____ Title Injury in Western Australia: The Health System Costs of...
Injury in Western Australia
The Health System Costs of Falls
in Older Adults
in Western Australia
Produced by Injury Prevention UnitPopulation Health Division© Department of Health
HP 1695
Injury in Western Australia:
The Health System Costs of Falls
in Older Adults in Western
Australia
Funded by: Report prepared by:
INJURY PREVENTION BRANCH INJURY RESEARCH CENTRE
Department of Health WA School of Population Health
189 Royal Street The University of Western Australia
East Perth WA 6004 35 Stirling Highway
Crawley WA 6009
Hendrie, D., Hall, S.E., Legge, M. & Arena, G.
November 2003
i
Injury in Western Australia: The Health System Cost of Falls in Older Adultsin Western Australia
ACKNOWLEDGMENTS
This research is a joint project of the Injury Prevention Branch of the Department of Health WesternAustralia and the Injury Research Centre of the University of Western Australia. This research wasundertaken during 2003 under contract to the Injury Research Centre, School of Population Health,University of Western Australia.
The authors would like to thank the following people and groups who assisted with this study –
• The Western Australian Linked Data Project at the Health Department Western Australia forproviding the data on hospital admissions.
• The Echo Project, Emergency Medicine, The University of Western Australia for providing the dataon emergency department presentations.
• Alison Lunt, Project Officer, National Demonstration Hospitals Program, Armadale Health Servicefor information used to validate assumptions for calculating the resource utilisation attributable t ofalls injuries.
• Associate Professor Michael Hobbs for his guidance in working with the hospital morbidity records.
Suggested Citation
Hendrie, D., Hall, S.E., Legge, M. and Arena, G. (2003) Injury in Western Australia: The Health SystemCost of Falls in Older Adults in Western Australia. Perth, Western Australia: Western AustralianGovernment.
Any enquiries or comments on this publication should be directed to:
Nicole Bennett, Injury Prevention Branch, Department of Health, Perth, Western Australia.E-mail: [email protected]
Delia Hendrie, Injury Research Centre, School of Population Health, University of Western Australia,Perth, Western Australia.E-mail: [email protected]
© Government of Western Australia 2003
There is no objection to this publication being copied in whole or part, provided there is dueacknowledgment of any material quoted from this publication.
ISBN: 1 876999 43 8
This report is available at:Injury Prevention Branch web-site:http://www.population.health.wa.gov.au/Promotion/resources_promotion.cfm#injury
OR
Injury Research Centre web-site:http://www.irc.uwa.edu.au/
Published by the Western Australian GovernmentINJURY RESEARCH CENTRE
DOCUMENT RETRIEVAL INFORMATION___________________________________________________________________________
Report No. Project No. Date Pages ISBNRR143 03-14 November 2003 26 1 876999 43 8
ii
___________________________________________________________________________Title
Injury in Western Australia: The Health System Costs of Falls in Older Adults in Western Australia___________________________________________________________________________Authors
Hendrie, D, Hall, SE*, Legge, M, Arena, G.**School of Population Health, The University of Western Australia
___________________________________________________________________________Performing Organisation
Injury Research CentreSchool of Population HealthThe University of Western Australia35 Stirling HighwayCRAWLEY WA 6009
Tel: (08) 9380 1302Fax: (08) 9380 1199___________________________________________________________________________Sponsor
Department of Health, Western Australia189 Royal StreetEast Perth WA 6004___________________________________________________________________________Abstract
The objective of this study was to determine the health system costs of falls in the population aged 65 years andabove in Western Australia for 2001/02. A population-based study was undertaken to calculate the health systemcost associated with older adults who presented to an Emergency Department (ED) in Western Australia. Thedata relating to the number of ED presentations and hospital admissions was obtained from hospital records.The use, and corresponding costs, of other health services such as outpatient services, medical services,community care, residential care and ancillary services had to estimated from alternative sources. There were 18706 ED attendances and 5 923 admissions to hospital for falls-related injuries in people aged 65 years andabove. The rates of ED attendance and hospital admission were higher for females than males, and increasedmore than ten fold in the 85 years and above age group compared with the 65-69 year age group. These falls costthe health system $83 million, with more than half of this attributable to hospital inpatient treatment.Assuming no changes in the current rate of falls by age group, the projected health system costs of falls in olderadults will increase to $174 million in 2021. Falls in older adults impose a substantial burden on healthservices, and a long-term strategic approach to falls prevention needs to be adopted. Policy in this area should betargeted at both reducing the current rate of falls through preventing injury in people from high risk groups andreducing the future rate of falls through generating a low risk population.___________________________________________________________________________KeywordsFalls, falls injuries, older adults, health system costs, community care costs___________________________________________________________________________
iii
TABLE OF CONTENTS
ACKNOWLEDGMENTS i
LIST OF TABLES iii
EXECUTIVE SUMMARY v
1. INTRODUCTION ................................................................................................................1
2. METHODS............................................................................................................................4
2.1 Study Design and Data Sources for Obtaining Cases...................................................4
2.2 Case Selection...............................................................................................................6
2.3 Valuing and Measuring the Use of Health System Resources......................................7
2.3.1 Assigning Costs to Health System Resources ....................................................72.3.2 Measuring the Volume of Resource Use...........................................................11
2.4 Data Analysis Issues..................................................................................................14
3. RESULTS.............................................................................................................................15
3.1 Number of Cases and the Use of Health System Resources......................................15
3.2 Health System Costs of Falls in Older Adults...........................................................19
3.3 Projections of Resource Utilisation and Health System Costs of Falls Injuries inOlder Adults in 2021..................................................................................................21
3.4 Effects on Resource Utilisation and Costs of Using Alternative Definitions ofa Falls-related Injury...................................................................................................24
4. DISCUSSION ......................................................................................................................26
REFERENCES ..............................................................................................................................28
iv
LIST OF TABLES
Table 2.1 Costing of Health System Resources.................................................................11
Table 3.1 Number and Rate per 10 000 Population of ED Presentations by OlderAdults, 2001/02 .................................................................................................16
Table 3.2 Number and Rate per 10 000 Population of Falls-Related HospitalAdmissions of Older Adults, 2001/02...............................................................17
Table 3.3 Other Resource Utilisation Resulting from Falls in Older People, 2001/02......18
Table 3.4 Total Health System Costs of Falls Injuries of Older Adults, 2001/02 ............20
Table 3.5 Composition of Health System Costs by Type of Cost, 2001/02....................20
Table 3.6 Projections of Hospital Care for Older Fallers, 2021 ........................................23
Table 3.7 Health System Cost of Falls of Older Adults, 2021..........................................23
Table 3.8 Effects on Resource Utilisation and Health System Costs of AlternativeDefinitions of a Falls-related Injury...................................................................25
v
EXECUTIVE SUMMARY
Objective
The objective of this study was to determine the health system costs of falls in the population
aged 65 years and above in Western Australia for 2001/02.
Methods
A population-based study was undertaken to calculate the health system costs of falls in older
adults who presented to any Emergency Department (ED) in Western Australia. The Emergency
Department Incident System (EDIS) was used to ascertain the number of people 65 years and
above who presented to an ED with a fall. The Injury Cost Database and the Western Australian
Linked Data Project were used to identify cases who were admitted to hospital with falls-related
injuries. The individual follow-up of subjects after discharge from the ED or hospital was beyond
the scope of the study, thus the use of health services other than ED and hospital inpatient care
had to be estimated from alternative sources. These health services included hospital care, medical
services, community care, residential care and ancillary services. A cost per unit was attached to
each of the resources used by health services to treat falls injuries and the volume of resources
used to treat older adults who had fallen was calculated. For each of the health services, the total
amount spent on falls injuries was estimated by multiplying the volume of resources used by
older fallers by the cost per unit of each resource. Age and gender specific rates of ED
presentations and hospital admissions were calculated based on the Western Australian
population of 65 years and above. Projections of future health service utilisation and
corresponding costs for 2021 were made based on population projections for Western Australia.
Results
Emergency Department presentations
• In 2001/02 the number of falls-related attendances at EDs in Western Australia by people
aged 65 years and above was 18 706.
v
• The rate of ED presentations for falls in older adults was 892 per 10 000 population. The
rate increased more than ten fold from 286 per 10 000 population in the 65 to 69 year age
group to 3287 per 10 000 population in the 85 years and above age group.
The rate for females was more than twice the rate for males at 1130 and 590 per 10 000
population respectively.
Hospital admissions
• In 2001/02 the number of hospital admissions of people aged 65 years and above for falls-
related injuries was 5923. Some people had more than one admission during 2001/02, with
4162 people accounting for the 5923 admissions.
• The rate of hospital admission for falls in people aged 65 years and above was 283 per
10 000 population. The rate for females was more than double the rate for males at 377 per
10 000 population and 165 per 10 000 population respectively.
• Increasing age was associated with increased likelihood of hospital admission with the rate
rising from 71 per 10 000 population in the 65 to 69 year age group to 1056 per 10 000
population in the 85 years and above age group.
Other resource utilisation
• In addition to ED presentations and hospital admissions, the use of health system
resources by people aged 65 years and above for falls injuries was estimated as:
– 9 300 ambulance trips, 106 500 allied health consultations; 85 100 personal care
assistant visits; 71 600 GP consultations; 59 000 hospital outpatient services; 31 300
nurse visits; 28 700 specialist consultations; 70 300 other domiciliary services; 200
high level residential care beds; and 50 000 prescriptions.
Health system costs
• The health system costs of falls injuries for people aged 65 years and above were
$83 million in 2001/02. This accounted for approximately 1.5% of health expenditure in
Western Australia.
vi
• The average health system cost per ED presentation was $4450, increasing from $3060 in
the 65 to 69 year age group to $5030 in the 80 to 84 year age group and then dropping to
$4680 in the 85 year and above age group. However, some patients have more than one
presentation for the same fall. Based on an estimate of the actual number of fallers, the
average health system costs per falls injury episode was $6500.
• The main components of the cost of falls to the health system were hospital inpatient
treatment (52%), high-level residential care (9%), hospital outpatient services (8%), ED
presentations (6%); and allied health consultations (5%).
Projections of resource utilisation and health system costs• Assuming no changes in the current rate of falls by age group and gender:
– The number of people aged 65 years and above presenting at an ED in Western
Australia in 2021 with a falls injury will double to 39 000.
– The number of hospital admissions and hospital bed days will also double to 12 300
and 157 300 respectively, assuming no change to hospital admissions policy and
lengths of stay. The additional bed days translate to an additional 220 bed acute care
facility that will be required to accommodate the demographic changes in the
population.
• Furthermore, assuming no changes in treatment patterns and costs:
– The projected health system costs of falls in older adults in Western Australia will be
$174 million in 2021. The corresponding costs in 2001/02 were $83 million.
Alternative definitions of a falls-related injury
• The definition of a falls-related hospital admission adopted in this study was that the
admission must have an external cause of ‘accidental fall’ and a principle diagnosis of an
injury or care involving use of rehabilitation procedures.
• Alternative definitions of what comprises a falls-related injury lead to different estimates of
the number of acute admissions for falls, which affects the resource utilisation and health
system costs attributable to falls in older adults.
viii
Discussion
In calculating the costs of falls-related injuries in people aged 65 years and above, the study
included only health system costs. Other costs such as the economic impact of falls to other
sectors and the health-related loss of quality of life that can result from a falls injury were not
considered. This implies that the estimated cost of falls injuries of $83 million is an underestimate
of the true burden to the community of falls in older people.
The cost of falls in older adults is a matter of increasing concern to public health practitioners and
planners, and these costs will escalate over the next 20 years unless effective prevention
programs are implemented. A need exists for a long-term strategic approach to be adopted in
relation to falls prevention, which is coherent and well coordinated.
Policy relating to falls prevention should be targeted at both reducing the current rate of falls
through preventing injury in people from high risk groups and reducing the future rate of falls
through generating a low risk population. The former requires the implementation of effective
falls prevention programs, while the latter can be achieved by targeting some of the underlying
risk factors of people who fall.
This study provides useful information on the potential cost savings that can be obtained if falls
prevention programs are successfully implemented. Alternatively, it highlights the cost outlays
needed if successful falls prevention programs are not implemented. This data can be used to
assist in the process of planning, prioritising and resource allocation in injury prevention
programs.
The information presented in this study is also useful for planning purposes, both in the short
term to use to identify the volume of services to be provided to treat falls injuries in older adults
and the costs of providing these services, and in the longer term to use in preparing projections of
future resource utilisation and budgetary costs. In addition to policymakers in falls prevention,
stakeholders who represent the interests of older people may find the incidence and cost data
ix
useful to understand how a proposed policy will impact on its target population and the problem
more generally.
In order to obtain a better understanding of the falls problem and its resource implications, the
data presented in the Injury Cost Database needs to be analysed in more depth. The geographical
distribution of falls incidence and its cost and the breakdown of falls and falls costs by type have
not been examined in this study, nor has the incidence of falls been compared with other
Australian states and territories to determine the relative magnitude of the falls problem across
jurisdictions. Also, the projections of resource implications and health system costs of falls
injuries in older adults were made using fairly simplistic assumptions such as the age-specific rate
of falls by gender remain constant over time to 2021. More sophisticated projections of the
incidence and cost of falls are needed to take account of the likelihood of a healthier older
population.
1
1. INTRODUCTION
Only a few studies have examined the costs of falls in older adults in Australia. These
studies have varied in their scope and the types of costs included, thus provide quite
different estimates of the costs of falls injuries in people aged 65 years and older.
Fildes et al. (1994) calculated the cost of falls in Australia for 1990. Using 1988
Australian data on the number of falls and 1991 Australian costs derived from U.S.
injury costs (Rice et al., 1989 cited in Fildes et al.), Fildes et al. estimated a cost of falls
of $1 954 million, (1990) assuming that Australian and US relative figures are the same,
i.e. the per head of population falling, sustaining an injury of a particular type and
severity, and the percentage resulting in death. This figure was not separated into age
groups so included younger fallers as well. The costs were calculated using an incidence-
based approach, which assesses the lifetime cost of injuries sustained in a given year.
Mathers and Penn (1999) calculated the health system costs of falls injuries in Australia
in 1993/94 as part of a study estimating the direct costs of health services attributable to
a comprehensive range of diseases and injury types. In contrast to Fildes et al. (1994),
this study used a prevalence-based approach to costing, which calculates the total cost
of injury in a given year, regardless of onset or when the injuries were sustained.
Mathers and Penn adopted a ‘top-down’ approach to costing that took known aggregate
expenditures on health care and apportioned these to disease and injury categories using
various sources of morbidity data. The direct health system cost of falls was estimated
as $471.3 million, of which $406.4 million was associated with falls in the elderly.
Watson and Ozanne-Smith (1997) calculated the lifetime costs of injury in Victoria for
1993/94 using a ‘bottom-up’ approach in which they collected data on the number and
type of injuries in Victoria and then estimated the aggregate direct and indirect costs
attributable to these injuries using an incidence-based approach. Direct costs consisted
of the cost of treating injury, which included the cost of hospital treatment, medical
consultations, pharmaceuticals, rehabilitation, attendant care and nursing home care.
2
Indirect costs included the value of lost output due to reduced productivity caused by
injury and any resultant disability and losses due to premature death.
The lifetime costs of falls in Victoria in 1993/94 were estimated as $557.3 million, of
which 49% was direct costs. For the 65 years and above age category, the lifetime costs
of falls injuries were estimated as $199.3 million.
Moller (1999) used the cost estimates produced by Watson and Ozanne-Smith together
with Australian figures on the incidence of injury to estimate the total lifetime costs of
injury by cause in Australia for 1995/96. The lifetime costs of falls was estimated as
$3 017 million, and the lifetime costs of falls in persons 65 years and older as $1 083
million. Subsequent work on the projected health system costs related to falls suggested
a cost in 2001 of $498 million for Australia and $42 million for Western Australia
(Moller 2003). These cost projections were prevalence-based, and derived from the
population estimates produced by the Australian Bureau of Statistics and the age
specific health system costs for falls injuries calculated by Mathers and Penn (1999).
Hall and Hendrie (2003) also produced prevalence-based costs of falls injuries for
Western Australia. Their study adopted a ‘bottom-up’ approach in which hospital and
three-month post-hospital community and personal costs associated with falls
sustained by older adults living in the community were calculated. Seventy-nine patients
participated in the study, which was conducted in 1999. The mean cost per patient was
between $4291 and $4642 (depending on assumptions used), which was extrapolated to
the Western Australian and Australian populations to provide a state and national figure
of $21 million and $287 million respectively for the cost of hospital and three-month
post-hospital community and personal care if those in residential care before the fall
were excluded and $48 million in Western Australia if they were included.
The objective of this study was to determine the health system costs of falls in the
population aged 65 years and above in Western Australia for 2001/02. Health system
costs included hospital treatment, ambulance transport, medical and allied health
3
consultations, community nursing, domiciliary services, equipment and home
modifications, pharmaceuticals, diagnostic tests and residential care. A prevalence-based
approach was adopted so costs reflect the burden to the health system of falls in a given
year.
Section 2 of this report describes the methods used in producing the data on the
incidence and costs of falls, Section 3 presents the results of the study, and Section 4
presents a discussion of the results and related issues.
4
2. METHODS
This section describes the methods used for this study under four main headings: study
design and data sources for obtaining cases, case selection, valuing and measuring the use
of health system resources and data analysis issues.
2.1 Study Design and Data Sources for Obtaining Cases
This was a population-based study to estimate the health system costs resulting from
falls in older adults who presented to any Emergency Department (ED) in Western
Australia.
Two distinct groups of cases were included in the study:
(1) Cases that presented to ED with an injury resulting from a fall and had one or more
hospital admission(s) associated with their fall, including transfers to other health
care providers or re-admissions. This group includes those admitted to the General
Holding Ward at ED.
(2) Cases that presented to ED and had no hospital admission associated with the fall
at that presentation.
Data for cases in the first group was available on an individual basis. The Injury Cost
Database and the Western Australian Linked Data Project were used to retrieve de-
identified data of falls injuries for the financial year 2001/2002. The linked data included
a string of all hospital morbidity records for the subject in order for hospital transfers
and re-admissions to be tracked, and the admission(s) related to the fall(s) to be
identified. The data included demographic variables (age, gender, postcode), hospital-
related variables (date of admission and discharge, length of stay, where admitted from
and discharged to, hospital location) and injury-related variables (diagnoses codes,
external causes of injury, diagnostic related group (DRG) category).
5
Individual records for cases in the second group, who only had an ED presentation,
were not accessed in this study.
6
For this group the Emergency Department Incident System (EDIS) was used to
ascertain the number of people who (i) presented to ED with a fall and met the inclusion
criteria for the study and (ii) were transferred, discharged home or to residential care, or
died within the ED.
The cases in this group did not generate a hospital morbidity record from the West
Australian Linked Data Project. EDIS contains data for the teaching hospitals and three
metropolitan non-teaching hospitals (Swan, Rockingham/Kwinana and
Armadale/Kelmscott). The number and characteristics of ED presentations for these
hospitals were extrapolated to all hospitals in Western Australia on the basis of the
proportion of ED attendances in Western Australia presenting at EDIS hospitals.
2.2 Case Selection
Cases for this study had to meet the following inclusion criteria:
• Be 65 years of age or above.
• Present to any ED in Western Australia after a fall.
• Present between 1 July 2001 and 30 June 2002 inclusive.
Cases, who had a hospital admission, also had to have:
• An external cause of ‘accidental fall’ (external cause code between W00.0 and
W19.9).
• A principle diagnosis of an injury (diagnosis code between S00.0 and T98.3) or care
involving use of rehabilitation procedures (Z50).
The impact on the number of cases and associated resource use and costs of adopting
alternative definitions of a fall was examined by changing the last inclusion criterion for
the case selection to (i) exclude cases who had a principle diagnosis of care involving use
of rehabilitation procedures (ii) include cases with at least one injury code in any of the
7
diagnosis fields and (iii) include all cases regardless of diagnosis codes (i.e. all cases with
an external cause of injury code indicating a falls-related admission).
2.3 Valuing and Measuring the Use of Health System Resources
2.3.1 Assigning Costs to Health System Resources
A cost was attached to each item identified as a health system resource.
Inpatient hospital episodes
For the group of cases admitted to hospital, the inpatient hospital cost for the
admission (including the ED component) was based on the Australian Related-
Diagnostic Related Group (AR-DRG) code recorded in their hospital morbidity record
for the fall-related admission and for any fall-related subsequent admissions within
2001/02. AR-DRG costs for AR-DRG Version 4.1 were available for teaching, other
public sector and private sector hospitals for 2000/01 (Commonwealth Department of
Health & Aged Care 2000). These were adjusted to 2001/02 prices using the hospital
price index produced by the Australian Institute of Health and Welfare (AIHW) (AIHW
2002). AR-DRG costs were assigned to each hospital admission on the basis of the
DRG code and type of hospital.
Emergency Department presentations
Cases that only attended the ED were assigned the average cost of an ED attendance for
Royal Perth Hospital (RPH) and Sir Charles Gairdner Hospital (SCGH) if they
attended an ED at a teaching hospital and 80% of this amount if they attended any other
ED. The average cost of an ED attendance for RPH and SCGH was $341 in 2002/03,
and this was adjusted to 2001/02 dollars ($328) using the AIHW’s total health price
index (personal communication1; AIHW 2002). Cases that were admitted to the General
Holding Ward of the ED or to the hospital have the cost of their ED attendance included
in the cost of their hospital admission.
Non-inpatient occasions of service
1 Personal communication with Finance Department, Royal Perth Hospital
8
Non-inpatient occasions of service included outpatient and plaster clinic visits. The
most recent published cost for these items was $109 for 1998/99, which was adjusted to
2001/02 prices using AIHW’s total health price index (Metropolitan Health Service
Board 1999; AIHW 2002).
Ambulance transport
The cost of ambulance transport to an ED was based on current costs adjusted to
2001/02 dollars using the AIHW’s total health price index (personal communication2;
AIHW 2002). The adjusted cost for 2001/02 of an emergency ambulance trip to hospital
was $437 per trip and a non-emergency trip from hospital to home or residential care
was $248 per trip.
General practitioner consultations
The schedule fee of a Level ‘B’ surgery consultation in the Commonwealth Medicare
Benefits Schedule was used as the cost of a general practitioner visit. A level ‘B’
consultation is a visit of less than 20 minutes, and the fee was $29.45 (Department of
Health and Ageing 2002). Any gap payable by the patient was not included.
Specialist consultations
The schedule fees for a consultant physician referred consultation in the Commonwealth
Medicare Benefits Schedule were used to cost specialist visits (Department of Health
and Ageing 2002). Using these fees a cost of $113.20 per visit was calculated based on
the weighted average of the schedule fees for an initial attendance at the surgery
($122.35), an initial attendance at home ($148.50) and a subsequent attendance at the
surgery ($61.25). The weights used to calculate this average were obtained from Hall
and Hendrie (2003). Again, any gap payable by the patient was not included.
Allied health consultations
The weighted average of the cost of all allied health services was calculated from Hall
and Hendrie (2003) as $38.00 per consultation for 1998/99, which was adjusted to
2 Personal communication with Finance Department, St John's Ambulance Service, WA
9
2001/02 prices ($40.60) using the AIHW total health price index (AIHW 2002). The
most commonly used allied health services after a fall were physiotherapy and then
occupational therapy.
10
Community nursing and domiciliary services
Community nursing is supplied by registered nurses. Domiciliary services included
personal care provided by personal care assistants and domestic services comprising
meals, home help and gardening. The cost of community nursing and domiciliary
services was calculated from Hall and Hendrie (2003) for 1989/99 and adjusted to
2001/02 prices using the AIHW total health price index (AIHW 2002). The 2001/02
costs were as follows: registered nurse, $60.40; personal care assistant, $27.80; a meal,
$5.30; a home help visit, $14.90; a gardening service visit; $14.90.
Equipment and home modifications
The costs of equipment and home modifications were based on Hall and Hendrie (2003).
Total expenditure on equipment and home modification was divided by the number of
participants in the study to obtain an average cost for these components of $65.00 per
faller expressed in 2001/02 prices (price adjustment using total health price index,
AIHW (2002)). Equipment included shower chairs, over-the-toilet seats, commodes and
walking aids. Home modifications included the installation of handrails or flexible
shower hoses.
Pharmaceuticals
The average cost of a prescription was obtained from Hall and Hendrie (2003), based on
the mix of pharmaceuticals used by participants in their study and the prices in the
Pharmaceutical Benefits Schedule. The average cost was $7.70 per prescription
expressed in 2001/02 prices (price adjustment using total health price index, AIHW
(2002)). The most commonly used pharmaceuticals were paracetamol, Panadeine and
anti-inflammatories.
Diagnostic tests
The cost of diagnostic tests was derived from Hall and Hendrie (2003). Total
expenditure on diagnostic tests was divided by the number of participants in the study
to obtain an average cost of $1.34 per faller expressed in 2001/02 prices (price
11
adjustment using total health price index, AIHW (2002)). Diagnostic tests included
blood tests and home visits for phlebotomy.
Residential care
The annual cost of a high-level residential care bed was $38 685 in 2001/02
(Productivity Commission 2003).
Table 2.1 summarises the data used to value the costs of health system resources.
Table 2.1 Costing of Health System Resources
Item Unit costs 2001/02 Source
Hospital inpatient episodes Variable depending on DRGcode and hospital type
Department of Health and AgeingAN-DRG cost weights (2000)
Emergency departmentpresentations
$328 for a teaching hospital;otherwise $262
Personal communication
Hospital outpatient services $109 per occasion of service Metropolitan Health Service Board(1999)
Ambulance transport $437 per trip for emergencytransport and $248 per trip fornon-emergency transport
Personal communication
General practitionerconsultations
$29.45 per visit Medicare Benefit Fee Schedule(2002)
Specialist consultations $113.20 per visit Medicare Benefit Fee Schedule(2002)
Allied health consultations $38.00 per visit Hall and Hendrie (2003)Community nursing anddomiciliary services
$60.40 per visit for aregistered nurse; $27.80 pervisit for a personal careassistant; $5.30 per meal;$14.90 per home help andgarden service visit
Hall and Hendrie (2003)
Equipment and homemodifications
Average cost of $65.00 perfaller
Hall and Hendrie (2003)
Pharmaceuticals Average cost of $7.70 perprescription
Hall and Hendrie (2003)
Diagnostic tests Average cost of $1.34 perfaller
Hall and Hendrie (2003)
High-level residential carebeds
$38 685 per bed per annum Productivity Commission (2003)
2.3.2 Measuring the Volume of Resource Use
In order to calculate the total cost for each health service, the volume of resource use had
to be calculated.
12
Inpatient hospital episodes
Individual records of each hospital admission that satisfied the inclusion criteria for this
study were available from the Injury Cost Database and the West Australian Linked
Data Project. This provided the number of falls-related hospital admissions.
Emergency Department presentations
The number of ED presentations was obtained from EDIS. As discussed in Section 2.1,
the number and characteristics of ED presentations for those hospitals included in EDIS
were extrapolated to all hospitals in Western Australia based on the proportion of ED
attendances in Western Australia presenting at EDIS hospitals.
Ambulance transport
The number of ambulance trips was calculated separately for cases living in the
community and those living in residential care facilities. Forty eight per cent of cases
who presented at an ED were assumed to require ambulance transport to the ED and
none were assumed to require ambulance transport from the hospital (Hall and Hendrie
2003). All cases living in residential care facilities were assumed to require ambulance
transport to and from hospital.
Hospital outpatient services, general practitioner and specialist consultations, allied
health consultations and pharmaceuticals
The individual follow-up of subjects after discharge from the ED or hospital was
beyond the scope of this study, therefore the use of health services other than ED and
hospital inpatient care had to be estimated from alternative sources. Mathers and Penn
(1999) provided Australian figures of health service utilisation relating to accidental falls
by age group for the following items: hospital admissions, non-inpatient occasions of
service, GP services, specialist services, prescriptions and allied health consultations.
This study assumed that the relationship between the number of each of the health
services utilised by fallers of 65 years and older and the number of falls-related hospital
admissions for this group had remained constant between 1993/94 and 2001/02 and was
13
the same for Western Australia as Australia. The Mathers and Penn ratio of the number
of each health service utilised to the number of falls-related hospital admissions was
calculated for each health service for the 65 years and older group by gender.
These ratios were then multiplied by the number of falls-related hospital admissions by
gender in 2001/02 to obtain an estimate of the volume of each of the health services used
by fallers of 65 years and above.
Community nursing and domiciliary services
Hall and Hendrie (2003) provide the only published figures for the utilisation of
community nursing and domiciliary services for fallers aged 65 years and over. The
average number of community services provided and domiciliary services used was
calculated for their study population of 79 fallers. Each case who presented at the ED,
excluding those in residential care, was assumed to utilise this average number of
community nursing and domiciliary services.
Equipment and home modifications and diagnostic tests
Again Hall and Hendrie (2003) provide the only published figures of the amount of
equipment and home modifications required by older fallers and the number of
diagnostic tests needed. No actual utilisation of these items was calculated for this
study. Rather the average cost of equipment and home modifications and diagnostic
tests was assigned to each case who presented at the ED.
Residential care
The Mathers and Penn (1999) study was also used as the basis for calculating the
number of high-level residential care beds attributable to falls. The assumption was
made that Mathers and Penn’s ratio of (i) the number of beds in high-level residential
care facilities attributable to falls to (ii) the Australian population of 65 years and above
had remained constant between 1993/94 and 2001/02 and that this ratio was the same
for Western Australia as for Australia. The Mathers and Penn ratio of the number of
beds in high level residential care facilities attributable to falls in older adults in
14
Australian in 1993/94 was applied to the Western Australian population of 65 years
and above in June 2002 to obtain an estimate of the number of high-level residential beds
attributable to falls in Western Australia in 2001/02 (ABS 2002). This calculation was
made separately for males and females.
2.4 Data Analysis Issues
The hospital morbidity records for the cases were identified and analysed using SAS
Version 8.2 (SAS Institute Inc. 2003). The data relating to ED presentations was
obtained from EDIS in tabular form. The volume of all items of health services included
as a health system cost was entered into an Excel spreadsheet for analysis. Total costs
for each health service were estimated by multiplying the volume of health services used
by the cost per unit of each service (Table 2.1).
Age and gender specific rates of ED presentations and hospital admissions were
calculated based on the Western Australian population of 65 years and above in June
2002 (ABS 2002). Projections of future health service utilisation and corresponding
costs for 2021 were made based on the ABS population projections. (ABS 2003) Series
B of the population projections was used; this series provides a mid-point estimate of
likely population changes. The assumptions used in making these projections was that
the current rate of falls by age group and gender remained constant between 2001/02 and
2021, no changes occurred in the treatment patterns of fallers and treatment costs
remained the same.
15
3. RESULTS
This section presents the number of cases included in the study, the use of health
system resources and the corresponding costs for falls-related injury in adults aged 65
years and over. The following data are presented –
(i) Number of cases and the use of health system resources.
(ii) Health system costs of falls injuries in older adults.
(iii) Projections of resource utilisation and health system costs of falls injuries in older
adults in 2021.
(iv) The effects on resource utilisation and costs of using alternative definitions of a
falls-related injury.
3.1 Number of Cases and the Use of Health System Resources
Emergency Department presentations
The number of falls-related attendances at an Emergency Department anywhere in
Western Australia in 2001/02 was estimated as 18 706 (Table 3.1). Females accounted
for 70% of fallers presenting at ED and the rate of falls presentations per 10 000
population increased with rising age.
As shown in Table 3.1, the rate of ED presentations for falls in people aged 65 years
and above was 892 per 10 000 population. The rate increased more than ten fold from
286 per 10 000 population in the 65 to 69 year age group to 3287 per 10 000
population in the 85 years and above age group. The rate for females was more than
twice the rate for males at 1130 and 598 per 10 000 population respectively. In each age
category, women were considerably more likely than men to fall and present at the ED.
The cumulative incidence is a measure of the proportion of new events or cases of a
disease or injury during a specified period of time. In 2001/02, the cumulative incidence
16
of falls requiring ED attendance in the population aged 65 years and above was 8.9% in
2001/02. This calculation is based on the number of ED presentations.
Each ED presentation does not represent a different person; the same person may
present twice at ED as a result of the same fall or may present twice at ED for separate
falls.
Table 3.1 Number and Rate per 10 000 Population of ED Presentations byOlder Adults, 2001/02
Age groups (years)
65-69 70-74 75-79 80-84 85+ Total
Males Number 820 918 1 082 1 073 1 719 5 612Rate per10000 popn
270 345 559 1 034 2 427 598
Females Number 940 1 379 2 148 2 741 5 886 13 094Rate per10000 popn
300 486 902 1 678 3 667 1 130
Total Number 1 760 2 297 3 230 3 814 7 606 18 706Rate per10000 popn
286 417 749 1 428 3 287 892
The three possible places of discharge from the ED are to return home (either their own,
family/friends, or residential care), be admitted to hospital or death. This study found
that approximately 70% of the fallers presenting at an ED were discharged home and the
remainder were admitted to hospital. Seven deaths of fallers were recorded in the EDIS
data as having occurred in an ED, which when extrapolated to all hospitals in Western
Australia suggested around 22 deaths in an ED at a state level.
Patients who remain in the General Holding ward of the ED generate a hospital
morbidity record and thus are counted as having a hospital admission and will have a
DRG based hospital cost the same as any other hospital admission. The discharge
destination of these patients is the same as for the ED only patients.
Inpatient hospital episodes
17
The number of hospital admissions of people aged 65 years and above for falls-related
injuries was 5923 in 2001/02 (Table 3.2). Some people had more than one admission
during 2001/02, with 4162 people accounting for the 5923 admissions.
Of the admissions, 79% had a primary diagnosis of an injury and the remaining cases
had a primary diagnosis of rehabilitation.
The majority of the hospital admissions were female (n=4 373, 74%). Tables 3.1 and
3.2 suggest that 33% of females presenting at an ED were admitted to hospital
compared with 28% of males. However, these percentages will overestimate the
admission rate as some rehabilitation cases will not have presented at an ED before
being admitted. The number of people admitted to hospital increased as age increased.
The rate of hospital admission for falls injuries in people aged 65 years and above was
283 per 10 000 population. The rate for females was more than double the rate for
males at 377 per 10 000 population and 165 per 10 000 population respectively.
Increasing age was associated with increased likelihood of hospital admission with the
rate rising from 71 per 10 000 population in the 65 to 69 year age group to 1056 per 10
000 population in the 85 years and above age group. The rate for females was almost
double the male rate in every age category except the youngest 65-69 year category.
Table 3.2 Number and Rate per 10 000 Population of Falls-Related HospitalAdmissions of Older Adults, 2001/02
Age group (years)
65-69 70-74 75-79 80-84 85+ Total
Males Number 180 235 322 316 497 1 550Rate per10000 popn
59 88 167 305 702 165
Females Number 256 435 743 993 1 946 4 373Rate per10000 popn
82 153 312 608 1212 377
Total Number 436 670 1 065 1 309 2 443 5 923Rate per10000 popn
71 122 247 490 1 056 283
18
The cumulative incidence of falls requiring hospital admission in the population 65
years and above was 2.8% in 2001/02.
While the hospital admissions data obtained from the Western Australian Linked Data
Project was linked to death records, a cause of death had not yet been attributed to all
deaths of cases.
Of the 4162 people who were hospitalised in 2001/02 with a falls injury, 909 had a link
to a death record but only 184 had a cause of death code.
Using the age-specific rates of death from accidental falls in people aged 65 years and
above produced by Cripps and Carman (2001), an estimated 99 people in Western
Australia would have died as a result of a fall in 2001/02.
Other resource utilisation
In addition to the emergency department and hospital inpatient resources utilised by
people aged 65 years and over after presentation at ED with a fall, a range of other
health system resources at outpatient departments or in the community are used in
providing treatment and long term care to this group. The estimated number of units of
these resources is shown in Table 3.3.
Table 3.3 Other Resource Utilisation Resulting from Falls in Older People,2001/02
Item Number
Male Female Total
Ambulance trips 2 800 6 500 9 300
Allied health consultations 42 600 63 900 106 500Personal care assistant visits 25 500 59 600 85 100GP consultations 19 100 52 500 71 600Hospital outpatient services 11 600 47 300 59 000Nurse visits 9 400 21 900 31 300Specialist consultations 6 600 22 100 28 700
19
Other domiciliary services 21 100 49 200 70 300High-level residential care beds 60 140 200
Prescriptions 15 400 34 500 50 000Equipment n.a. n.a. n.a.Lab tests n.a. n.a. n.a.
Almost half of the patients who attended the ED arrived by ambulance. Typically,
family and friends would have taken the other half to ED; a few may have driven
themselves or gone by taxi depending on their injuries and resources.
The two most commonly used professional services were those provided by allied
health professionals and by GPs. There were also a significant number of visits to
hospital outpatient services; these patients would have seen a variety of health care
providers in a variety of settings such as a plaster clinic after a fracture or for
rehabilitation.
Care provided by personal care assistants, mostly through community nursing care
agencies, was the second most commonly used service. Registered nurses would have
attended patients in need of higher-level services such as those requiring wound care or
assistance with medications.
Males were more likely to visit an allied health professional post-fall than females who
fell, whereas females were more likely to visit a specialist consultant or have a non-
inpatient service provided. This may be due to underlying patterns of injury, such as
the rate of fractures or to demographic characteristics such as age.
3.2 Health System Costs of Falls in Older Adults
The health system costs of accidental falls for people aged 65 years and above was
$83 million in 2001/02 (Table 3.4). This accounted for approximately 1.5% of all health
expenditure in Western Australia. The costs attributable to females was considerably
greater than to males, reflecting the female to male ratios in the number of fallers
presenting at ED and the associated higher number of hospital admissions. Similarly, the
20
costs rose as age increased, with the 85 years and above age group accounting for 43%
of health system expenditure on falls-related injuries.
The average health system cost per ED presentation was $4450, increasing from $3060
in the 65 to 69 year age group to $5030 in the 80 to 84 year age group and then
dropping to $4680 in the 85 year and above age group. However, some patients have
more than one ED presentation for the same fall. Based on an estimate of the actual
number of fallers, the average health system cost per falls injury episode was $6500.
This amount increased from $4560 in the 65 to 69 year age group to $7370 in the 80 to
84 year age group, then dropped to $6700 in the 85 year and above age group.
Table 3.4 Total Health System Costs of Falls Injuries of Older Adults,2001/02
Age group (years)
65-69 70-74 75-79 80-84 85+ Total
Males Cost ($000) 2 299 3 078 4 549 4 889 7 883 22 698% of total cost 2.8 3.7 5.5 5.9 9.5 27.3
Females Cost ($000) 3 093 5 229 10 216 14 290 27 712 60 541% of total cost 3.7 6.3 12.3 17.2 33.3 72.7
Total Cost ($000) 5 393 8 307 14 765 19 179 35 595 83 238% of total cost 6.5 10.0 17.7 23.0 42.8 100.0
The cost of hospital inpatient services was by far the greatest cost, accounting for 52%
of overall health system costs (Table 3.5). The next two largest cost items were high-
level residential care and hospital outpatient services. The share of residential care was
high despite the relatively few beds attributable to falls injuries because of the high
annual cost of a residential care bed.
Table 3.5 Composition of Health System Costs by Type of Cost, 2001/02
Item Cost
$000 %
21
Inpatient hospital treatment 43 400 52.1High-level residential care 7 900 9.4Hospital outpatient services 7 000 8.4Emergency department presentations 5 000 6.0Allied health consultations 4 300 5.2Ambulance transport 4 000 4.8Specialist consultations 3 200 3.9Domiciliary care services 2 800 3.4GP consultations 2 100 2.5Community nursing 1 900 2.3Other 1 200 1.5Pharmaceuticals 400 0.5Total 83 200 100.0
3.3 Projections of Resource Utilisation and Health System Costs of Falls Injuries in
Older Adults in 2021.
The population of Western Australia is aging and thus the number of falls in people
aged 65 years and over is expected to increase, unless there is an improvement in the
availability and effectiveness of prevention strategies and their uptake in the
community. This section provides estimates of the changes in the level of resources that
will be needed to respond to falls-related injuries and the associated costs. The
projections are for 2021 based on the assumption that the age-specific rate of falls by
gender remains constant over time to 2021. Further assumptions were that injury
patterns stays the same as at present, thus the proportions presenting at ED and
admitted to hospital stay constant, treatment policy relating to falls injuries does not
change, and the costs of treating falls remains the same.
Projected resource utilisation in 2021
The increased demand for health services to treat falls-related injuries is shown in Table
3.6. The number of people aged 65 years and above presenting at an ED with a falls
injury will double to 39 000. This will have a flow on effect with a doubling of the
number of falls related admissions and number of hospital bed days required to treat
falls patients. Assuming no changes to hospital admissions policy and length of stays,
431 beds will be occupied all year round by patients 65 years and above with a falls
injury. The additional bed days for treating falls injuries in 2021 (n=81 900) translate to
an additional 220 bed acute care facility that will be required to accommodate the
demographic changes in the population.
22
In addition, the increase in the number of hospital outpatient services will require
additional infrastructure and staff to support the extra 121 000 patient visits. There are
also resource implications for community care, residential care and ancillary care with
similar proportional increases in their service provisions.
23
Table 3.6 Projections of Hospital Care for Older Fallers, 2021
Item Projected number
Hospital inpatient treatment 12 300 - Bed days 157 300
Allied health consultations 228 500Personal care assistants 179 100GP consultations 149 200Domiciliary services 148 000Hospital outpatient services 121 000Prescriptions 104 900Community nurse visits 65 800Specialist consultations 59 300Emergency department 39 400Ambulance trips 19 600High-level residential care beds 400
Projections of the health system costs of falls in 2021
The substantial increases in the number of resources used in the hospital, community
and residential care sectors have major repercussions for the cost of falls in people aged
65 years and over in Western Australia in 2021. The projected health system costs of
falls in Western Australia in 2021 will be $174 million. The corresponding costs in
2001/02 were $83 million.
Table 3.7 Health System Cost of Falls of Older Adults, 2021
Age group (years)
65-69 70-74 75-79 80-84 85+ Total
Males Cost ($000) 5 025 6 815 9 244 11 687 22 923 55 694% of total cost 2.9 3.9 5.3 6.7 13.2 32.0
Females Cost ($000) 6 802 11 126 17 676 25 712 57 147 118 463% of total cost 3.9 6.4 10.1 14.8 32.8 68.0
Total Cost ($000) 11 827 17 941 26 919 37 399 80 070 174 157% of total cost 6.8 10.3 15.5 21.5 46.0 100.0
24
3.4 Effects on Resource Utilisation and Costs of Using Alternative Definitions of a
Falls-related Injury
The definition of a falls-related hospital admission adopted in this study was the
admission must have an external cause of ‘accidental fall’ and a principle diagnosis of an
injury or care involving use of rehabilitation procedures.
Using this definition the number of falls-related admissions was 5923. Excluding the
cases admitted for care involving use of rehabilitation procedures, the number of acute
admissions for falls would be 4693. Alternatively, widening the definition to include all
falls-related admissions with an injury diagnosis in any of the diagnosis fields would
result in 7303 admissions being falls-related, or 7320 if all admissions with an external
cause of ‘accidental fall’ were counted. Table 3.8 shows the effect on resource use and
health system costs of using alternative definitions of a fall. Health system costs range
from $70.3 million if only acute admissions are included to $100.4 million if all
‘accidental falls’ are counted. Similarly the projected health system costs for 2021 range
from $147.2 million to $210.8 million.
The purpose of this study was to determine the health system costs of falls in the adult
population aged 65 years and above, thus all acute and rehabilitation admissions were
included as falls-related. Clearly, widening the definition to include all admissions with
an injury diagnosis in any field would overestimate the number of falls-related hospital
episodes. The primary diagnoses of patients with an injury diagnosis in fields other than
the primary field include diseases of the circulatory, nervous and respiratory systems,
neoplasms, mental and behavioural disorders and diseases of the musculoskeletal system
and connective tissue. However, some resources used by this group would have been as
a result of their fall thus the estimates of the costs to the health system of falls in older
adults produced in this study are conservative.
25
Table 3.8 Effects on Resource Utilisation and Health System Costs of
Alternative Definitions of a Falls-related Injury
Alternative definitions Number ofhospital
admissions
Health systemcosts 2001/02
($000)
Projected healthsystem costs2021 ($000)
Falls-related admission and injury asprimary diagnosis
4 693 70 333 147 197
Falls-related admission and injury orrehabilitation as primary diagnosis (basecase)
5 923 83 238 174 157
Falls-related admission and any diagnosisof an injury
7 303 100 129 210 133
Falls related admission 7 320 100 389 210 759
26
4. DISCUSSION
The purpose of this study was to determine the health system costs of falls in older
adults aged 65 years and above. These costs include hospital treatment, medical
services, community care, residential care and ancillary services. The population for
which costs were calculated was older adults who presented to any Emergency
Department in Western Australia.
The number of falls-related presentations at an ED anywhere in Western Australia in
2001/02 was 18 706 or 892 per 10 000 population. The number of hospital admissions
was 5 923. Depending on the type and severity of their injuries, these fallers had
contact with a range of different health services on discharge from ED or hospital. The
overall health system costs of accidental falls in older people across all these services
were estimated to be $83 million. With the future ageing of the population, the demand
for health services as a result of the increase in the number of falls injuries, in the
absence of effective prevention and lower treatment costs, was projected to be $174
million in 2021.
A strength of this study was it was population-based, with cases identified from ED
and hospital morbidity records. However, the individual follow-up of subjects after
discharge from the ED or hospital was beyond the scope of this study so the use of
health services other than ED and hospital inpatient care had to be estimated from
alternative sources. While this introduces some uncertainty into the results of the study,
the best possible known sources were used as the basis for determining utilisation of
health system resources by older fallers. In order to calculate the costs associated with
the utilisation of health system resources, a cost per unit had to be assigned to each item
of expenditure. Some items such as hospital inpatient care were easy to allocate a cost
as DRG cost weights provide a readily available source of the cost of hospital
admissions. However, a number of other items were more difficult to cost and, as was
the case with resource use, estimates of the cost per unit of these items were made
based on the best possible known sources.
27
Because of the perspective of the study, only costs to the health system were covered.
The economic impact of falls in older people to other sectors was not considered. In
addition, a fall can be a very significant event for an older person, resulting in a loss of
mobility and independence. Measured on an interval scale from between 0 (death) and 1
(full health), a ‘bad’ hip fracture (which results in admission to a nursing home) has been
valued at 0.05; a ‘good’ hip fracture (maintaining independent living in the community)
at 0.31, and the fear of falling at 0.67. Eighty percent of the women who provided these
valuations would rather be dead (utility=0) than experience the loss of independence and
quality of life that results from a bad hip fracture and subsequent admission to a nursing
home (Salkeld et al. 2000). This study has not included the non-monetary cost of loss of
quality of life associated with falls injuries in older adults.
Of the studies that have examined the costs of falls in older adults in Australia, the most
relevant ones with which to compare the results of this study are Moller (2003) and
Hall and Hendrie (2003) since both provided prevalence-based figures of health system
costs. Mathers and Penn (1998) also provided prevalence-based figures of health
system costs; however, Moller has used Mathers and Penn’s estimates for 1993/94 as
the basis for his projections of health system costs in 2001 so the latter are more useful
for comparison purposes for this study. Other studies of the costs of falls injuries in
older adults in Australia have calculated incidence-based or lifetime costs, which cannot
be compared in order of magnitude to estimates derived using a prevalence based
approach (Fildes 1994); Watson and Ozanne-Smith 1997)
Moller’s estimate (2003) of the cost of falls in 2001 was $42 million, which is
considerably below the $83 million presented in this study. One reason for the
difference in costs between the two studies is that Moller did not take into account any
increase in health care costs since 1993/94. Adjusting for the increase in health system
costs raises the Moller estimate to $49 million. A second reason for the cost difference
is that Moller based his calculations on 3300 falls-related hospital admissions in
28
Western Australia in 2001 whereas the number using the definition adopted in this
study was 5923.
If Moller’s estimates are adjusted upwards proportionately based on the increased
number of hospital admissions (i.e. multiplying his estimate of $49 million by 5923
divided by 3300) then the Moller estimate of the cost of falls injuries in older adults
becomes $89 million, which is a similar order of magnitude to the findings of this study.
The extrapolation by Hall and Hendrie (2003) of the hospital and three-month post-
hospital community and personal costs associated with falls in older adults in Western
Australia was $48 million in 1999. Again, this is well below the $83 million estimated in
this study. The main reasons for this difference in costs is that their extrapolation was
based on 11 071 patients presenting at an ED in Western Australia. This number of ED
presentations was based on the total number of hospital admissions for falls in older
adults in Western Australia together with the finding for their study population that
53% of ED presentations were admitted to hospital. This admission rate is too high at a
state wide level. It was based on a teaching hospital, which would be likely to have a
higher admission rate than the average across all hospitals, and was derived from a
relatively small sample of patients. Using population-based data this study suggests an
admission rate of around 30%. Reworking Hall and Hendrie’s estimate based on this
lower admission rate provides a total cost of hospital and three-month post-hospital
community and personal costs of $85 million. Again this is very similar to the estimate
of $83 million found in this study. While the Hall and Hendrie estimate represents costs
to three months only, their study shows the use of health services reducing in
successive months after hospital discharge. In addition, Hall and Hendrie did not restrict
their costing to health system costs so their estimate includes a few costs to sectors
other than the health system.
The costs of falls in older people are a substantial burden on the health system and will
escalate over the next 20 years unless effective prevention programs are implemented.
While some falls prevention programs are currently in place, a need exists for a long-
29
term strategic approach to be adopted that is evidence based and coherent. Policy
relating to falls prevention programs should be targeted at both reducing the current rate
of falls through preventing injury in people from high risk groups and reducing the
future rate of falls through generating a low risk population.
The former requires the implementation of effective falls prevention programs, while the
latter can be achieved by targeting some of the underlying risk factors of people who
fall.
Many recent studies have evaluated the efficacy of falls prevention programs, and
several economic evaluations of falls prevention programs have been conducted. This
study provides useful information on the potential cost savings that can be obtained if
falls prevention programs are successfully implemented. Alternatively, it highlights the
cost outlays needed if successful falls prevention programs are not implemented. Given
evidence on the effectiveness of falls prevention programs, the cost of falls data from
this study can be used to calculate the break-even point of a falls prevention program,
which is the point at which program costs exactly equal the cost savings generated from
a reduction in the number of falls. Alternatively, the cost of falls data can be used to
calculate the cost-effectiveness of falls prevention programs measured in terms of the
cost per injury fall prevented. These types of analyses can assist in the process of
planning, prioritising and resource allocation in injury prevention programs.
The information presented in this study is also useful for planning purposes, both in the
short term to use to identify the volume of services to be provided to treat falls injuries
in older adults and the costs of providing these services, and in the longer term to use in
preparing projections of future resource utilisation and budgetary costs.
In addition, policymakers in falls prevention and stakeholders who represent the
interests of older people require information to understand how a proposed policy will
impact on its target population and the problem more generally. Incidence and cost data
plays a useful role in advising the variety of groups in the policy making process.
30
In order to obtain a better understanding of the falls problem and its resource
implications, the data presented in the Injury Cost Database needs to be analysed in
more depth. For example, this study has not addresses the geographical distribution of
falls incidence and its cost nor has it examined the breakdown of falls and falls costs by
type.
Furthermore, while the cost of falls in this study has been compared with that found in
other studies, no comparative analysis of the incidence of falls in other Australian states
and territories was made to determine the relative magnitude of the falls problem across
jurisdictions.
Finally, the projections of resource implications and health system costs of falls injuries
in older adults were made using fairly simplistic assumptions such as the age-specific
rate of falls by gender remain constant over time to 2021. More sophisticated
projections of the incidence and cost of falls are needed to take account of the likelihood
of a healthier older population.
31
REFERENCES
Australian Bureau of Statistics 2002, Australian Demographic Statistics, Cat. No.3101.0, ABS, Canberra.
Australian Bureau of Statistics 2003, Population Projections, Australia, 2002-2011, Cat.No. 9216.0, ABS, Canberra.
Australian Institute of Health & Welfare 2002, Health Expenditure, 2000-01, AIHWCat. No. HWE 20, Australian Institute of health and Welfare, Canberra.
Commonwealth Department of Health & Aged Care 2000, NHCDC Round 3 (1998-99):Final Cost Report, [Online], Available:http://www.health.gov.au/casemix/costing/fc_r3.htm [2003 13 Jul.].
Department of Health & Ageing 2002, Medicare Benefits Schedule, [Online], Available:http://www.health.gov.au/pubs/mbs/index.htm [2003, 11 Jul].
Fildes, B., Dunt, D., Oxley, J., Lee, S. & Day, L. 1994, ‘Analysis of elderly mortalitydue to injury’, in Injuries Among Older People, ed. B. Fildes, Collins Dove, Melbourne,pp. 55-88.
Hall, S.E. & Hendrie, D.V. 2003, ‘A prospective study of the costs of falls in olderadults living in the community’, Australian and New Zealand Journal of Public Health,vol. 27, no. 3, pp. 343-361.
Mathers, C. & Penn, R. 1999, Health System Costs of Injury, Poisoning andMusculoskeletal Disorders in Australia 1993-94, Australian Institute of Health andWelfare, Canberra.
Metropolitan Health Service Board 1999, Annual Report 1997-1998, MetropolitanHealth Service Board, Perth.
Moller, J. 1999, ‘The cost of injury’, Injury Issues Monitor, vol. 15, p. 4. Available:http://www.nisu.flinders.edu.au/pubs/injcost/ [2003, 11 Jul.].
Moller, J., 2003, Projected Costs of Fall Related Injury to Older Persons due toDemographic Change in Australia, New Directions in Health and Safety, Adelaide.
Productivity Commission 2003, Productivity Commission Submission to the Review ofPricing Arrangements in Residential Aged Care, Productivity Commission, Melbourne.
SAS Institute Inc. 2003, Carey, NC, USA.
32
Salkeld, G., Cameron, I.D., Cumming, R.G., Easter, S., Seymour, J., Kurrie, S.E. &Quine, S. 2000, ‘Quality of life related to fear of falling and hip fracture in older women:a time trade off study’, British Medical Journal, vol. 320, pp. 341-346.
Watson, W.L. & Ozanne-Smith, J. 1997, The Cost of Injury to Victoria, Report No. 124,Monash University Accident Research Centre, Melbourne.
Injury in Western Australia
The Health System Costs of Falls
in Older Adults
in Western Australia
Produced by Injury Prevention UnitPopulation Health Division© Department of Health
HP 1695
